Machine learning-assisted preparation of HMF and process optimization

CHEN Dayang; ZHANG Yibin; CHEN Yiming; DU Weichen; ZHAO Huaiyuan; HOU Zhaoyin

doi:10.11949/0438-1157.20260557

您当前的位置：

首页 >

文章列表页 >

Machine learning-assisted preparation of HMF and process optimization

更新时间：2026-05-25

- Machine learning-assisted preparation of HMF and process optimization
- CIESC Journal (2026)
- 作者机构：
  
  1.浙江大学化学系，浙江省全省高值化学品低碳合成重点实验室，浙江杭州 310058
  2.浙江恒逸石化研究院有限公司，浙江杭州 311200
- 作者简介：
- 基金信息：
- DOI：10.11949/0438-1157.20260557
  CLC： O643.3
- Received：18 April 2026，
  
  Revised：2026-05-25，
  
  Accepted：25 May 2026，
- 稿件说明：
移动端阅览
CHEN Dayang, ZHANG Yibin, CHEN Yiming, et al. Machine learning-assisted preparation of HMF and process optimization[J/OL]. CIESC Journal, 2026.
DOI：

CHEN Dayang, ZHANG Yibin, CHEN Yiming, et al. Machine learning-assisted preparation of HMF and process optimization[J/OL]. CIESC Journal, 2026. DOI： 10.11949/0438-1157.20260557.

摘要

高质量标准化数据集匮乏是制约催化反应机器学习和建模的主要障碍。以5-羟甲基糠醛(5-Hydroxymethylfurfural

HMF)的制备为目标，首先借助大语言模型的文本处理能力开展了大规模、选择性的文献数据提取；然后结合机器学习进行反应原料、溶剂及相关催化剂的筛选；并在机器学习的指导下有针对性地制备了具有多级孔结构的有机聚合物固体酸(

ierarchical porous

ulfonated

olymeric

MBP and

BSA

hSPBCH)催化剂；最后通过人工智能辅助进一步优化了反应的工艺条件以提升HMF的产率。机器学习的研究结果表明：以果糖为原料时得到的HMF平均产率最高，表明其在产率导向的HMF制备体系中具有明显优势；采用磺酸基固体酸催化剂及纯水溶剂，HMF的产率最高可达70.0%，引入有机溶剂后可将HMF产率进一步提高至79.2%。经过机器学习后的优化和实验验证，发现在自制的hSPBCH催化剂作用下、以果糖为原料、二甲基亚砜(dimethyl sulfoxide

DMSO)为溶剂、155℃下反应2h后，HMF产率提升至92.8%，HMF的时空产率最高可达45.6 g-HMF/g-cat/h。本研究提出的方法有助于推动机器学习在HMF制备及催化剂研制领域的发展与应用。

Abstract

The scarcity of high-quality standardized datasets is a major obstacle constraining machine learning and modeling in catalytic reactions. This study focuses on the preparation of HMF(5-hydroxymethylfurfural

HMF) via machine learning. First

leveraging the text-processing capabilities of large language models

large-scale and selective extraction of literature data was conducted. Subsequently

machine learning was employed to screen the raw materials

solvents

and related catalysts for HMF production. Based on the machine learning results

a hierarchically porous organic polymer solid acid catalyst (

ierarchical porous

ulfonated

olymeric

MBP and

BSA

hSPBCH) was purposefully prepared. Finally

artificial intelligence-assisted optimization of the reaction process conditions was carried out to enhance the yield of HMF. The machine learning results indicate that fructose as the raw material yields the highest average HMF production

indicating that it has a distinct advantage in yield-oriented HMF preparation. Using sulfonic acid-based solid acid catalysts and pure water as the solvent

the yield of HMF can reach 70.0%

while the introduction of organic solvents can increase the HMF yield to 79.2%. After machine learning-guided optimization and experimental validation

it was found that under the action of the self-prepared hSPBCH catalyst

using fructose as the raw material and dimethyl sulfoxide (DMSO) as the solvent

the yield of HMF w

as increased to 92.8% at 155°C

with the highest space-time yield of HMF reaching 45.6 g-HMF/g-cat/h. The methodology proposed in this study can contribute to promoting the development and application of machine learning in the field of HMF preparation and catalytic reactions.

关键词

Keywords

references

Nasrollahzadeh M , Bidgoli N S S , Shafiei N , et al . Biomass valorization: Sulfated lignin-catalyzed production of 5-hydroxymethylfurfural from fructose [J ] . International Journal of Biological Macromolecules , 2021 , 182 : 59 - 64 .

Gan L H , Lyu L , Shen T R , et al . Sulfonated lignin-derived ordered mesoporous carbon with highly selective and recyclable catalysis for the conversion of fructose into 5-hydroxymethylfurfural [J ] . Applied Catalysis A: General , 2019 , 574 : 132 - 143 .

Shimizu K I , Uozumi R , Satsuma A . Enhanced production of hydroxymethylfurfural from fructose with solid acid catalysts by simple water removal methods [J ] . Catalysis Communications , 2009 , 10 ( 14 ): 1849 - 1853 .

Wang S G , Zhang Z H , Liu B . Catalytic conversion of fructose and 5-hydroxymethylfurfural into 2, 5-furandicarboxylic acid over a recyclable Fe 3 O 4 –CoO x magnetite nanocatalyst [J ] . ACS Sustainable Chemistry & Engineering , 2015 , 3 ( 3 ): 406 - 412 .

Román-Leshkov Y , Chheda J N , Dumesic J A . Phase modifiers promote efficient production of hydroxymethylfurfural from fructose [J ] . Science , 2006 , 312 ( 5782 ): 1933 - 1937 .

Qu Y S , Huang C P , Zhang J , et al . Efficient dehydration of fructose to 5-hydroxymethylfurfural catalyzed by a recyclable sulfonated organic heteropolyacid salt [J ] . Bioresource Technology , 2012 , 106 : 170 - 172 .

Salak Asghari F , Yoshida H . Acid-catalyzed production of 5-hydroxymethyl furfural from d-fructose in subcritical water [J ] . Industrial & Engineering Chemistry Research , 2006 , 45 ( 7 ): 2163 - 2173 .

Qi X H , Watanabe M , Aida T M , et al . Catalytical conversion of fructose and glucose into 5-hydroxymethylfurfural in hot compressed water by microwave heating [J ] . Catalysis Communications , 2008 , 9 ( 13 ): 2244 - 2249 .

Teong S P , Yi G S , Zhang Y G . Hydroxymethylfurfural production from bioresources: past, present and future [J ] . Green Chemistry , 2014 , 16 ( 4 ): 2015 - 2026 .

Wang T F , Nolte M W , Shanks B H . Catalytic dehydration of C6 carbohydrates for the production of hydroxymethylfurfural (HMF) as a versatile platform chemical [J ] . Green Chemistry , 2014 , 16 ( 2 ): 548 - 572 .

Portillo Perez G , Mukherjee A , Dumont M J . Insights into HMF catalysis [J ] . Journal of Industrial and Engineering Chemistry , 2019 , 70 : 1 - 34 .

Li M F , Zhang Q T , Luo B , et al . Lignin-based carbon solid acid catalyst prepared for selectively converting fructose to 5-hydroxymethylfurfural [J ] . Industrial Crops and Products , 2020 , 145 : 111920 .

Jordan M I , Mitchell T M . Machine learning: Trends, perspectives, and prospects [J ] . Science , 2015 , 349 ( 6245 ): 255 - 260 .

Butler K T , Davies D W , Cartwright H , et al . Machine learning for molecular and materials science [J ] . Nature , 2018 , 559 ( 7715 ): 547 - 555 .

Raccuglia P , Elbert K C , Adler P D F , et al . Machine-learning-assisted materials discovery using failed experiments [J ] . Nature , 2016 , 533 ( 7601 ): 73 - 76 .

Janet J P , Kulik H J . Resolving transition metal chemical space: feature selection for machine learning and structure–property relationships [J ] . The Journal of Physical Chemistry A , 2017 , 121 ( 46 ): 8939 - 8954 .

Gómez-Bombarelli R , Wei J N , Duvenaud D , et al . Automatic chemical design using a data-driven continuous representation of molecules [J ] . ACS Central Science , 2018 , 4 ( 2 ): 268 - 276 .

Butler K T , Frost J M , Walsh A . Ferroelectric materials for solar energy conversion: photoferroics revisited [J ] . Energy & Environmental Science , 2015 , 8 ( 3 ): 838 - 848 .

Chen A , Zhang X , Chen L T , et al . A machine learning model on simple features for CO 2 reduction electrocatalysts [J ] . The Journal of Physical Chemistry C , 2020 , 124 ( 41 ): 22471 - 22478 .

Segler M H S , Preuss M , Waller M P . Planning chemical syntheses with deep neural networks and symbolic AI [J ] . Nature , 2018 , 555 ( 7698 ): 604 - 610 .

Coley C W , Green W H , Jensen K F . Machine learning in computer-aided synthesis planning [J ] . Accounts of Chemical Research , 2018 , 51 ( 5 ): 1281 - 1289 .

Brown T B , Mann B , Ryder N , et al . Language models are few-shot learners [C ] // Proceedings of the 34th International Conference on Neural Information Processing Systems . December 6 - 12 , 2020 , Vancouver, BC, Canada . ACM , 2020: 1877 - 1901 .

Devlin J , Chang M W , Lee K , et al . BERT: pre-training of deep bidirectional transformers for language understanding [C ] // Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) . Minneapolis, MinnesotaAssociation for Computational Linguistics , 2019 : 4171 - 4186 .

Krallinger M , Rabal O , Lourenço A , et al . Information retrieval and text mining technologies for chemistry [J ] . Chemical Reviews , 2017 , 117 ( 12 ): 7673 - 7761 .

Swain M C , Cole J M . ChemDataExtractor: a toolkit for automated extraction of chemical information from the scientific literature [J ] . Journal of Chemical Information and Modeling , 2016 , 56 ( 10 ): 1894 - 1904 .

Zhang J , Das A , Assary R S , et al . A combined experimental and computational study of the mechanism of fructose dehydration to 5-hydroxymethylfurfural in dimethylsulfoxide using Amberlyst 70, PO4 3−/niobic acid, or sulfuric acid catalysts [J ] . Applied Catalysis B: Environmental , 2016 , 181 : 874 - 887 .

Baruah J , Medhi U , Nath B K , et al . Enhancements in the production of 5-HMF from glucose: a review on recent advances in heterogeneous catalysts and solvent effects [J ] . Next Energy , 2025 , 9 : 100420 .

Liu T , Niu Y M , Cheng K X , et al . Exploring the formation pathway and antioxidant properties of the sugar-smoking pigment 5-GGMF [J ] . Food Chemistry , 2024 , 442 : 138406 .

Marianou A A , Michailof C M , Pineda A , et al . Effect of Lewis and Brønsted acidity on glucose conversion to 5-HMF and lactic acid in aqueous and organic media [J ] . Applied Catalysis A: General , 2018 , 555 : 75 - 87 .

Zhao Y , Lu K F , Xu H , et al . A critical review of recent advances in the production of furfural and 5-hydroxymethylfurfural from lignocellulosic biomass through homogeneous catalytic hydrothermal conversion [J ] . Renewable and Sustainable Energy Reviews , 2021 , 139 : 110706 .

董喜恩 , 蒋锋 , 罗根祥 . 糖类脱水生成5-羟甲基糠醛的研究进展 [J ] . 化工中间体 , 2010 , 6 ( 2 ): 17 - 22 .

Dong X E , Jiang F , Luo G X . Dehydration of Carbohydrates into 5-Hydroxymethylfurfural in the presence of double micro-porous zeolite catalyst [J ] . Chemical Intermediate , 2010 , 6 ( 2 ): 17 - 22 .

Ren Q H , Ma H , Wang W B , et al . Catalytic dehydration of fructose to 5-hydroxymethylfurfural over mesoporous Nb-W oxide solid acid catalyst [J ] . Biomass and Bioenergy , 2023 , 174 : 106860 .

Shi H , Ramasamy K K , Ma R S , et al . Nanoporous catalysts for biomass conversion [M ] // Nanoporous Materials for Molecule Separation and Conversion . Amsterdam : Elsevier , 2020 : 387 - 440 .

Jiang Y Y , Zhou R R , Ye B Y , et al . Acetalization of glycerol over sulfated UiO-66 under mild condition [J ] . Journal of Industrial and Engineering Chemistry , 2022 , 110 : 357 - 366 .

Sonsiam C , Kaewchada A , Pumrod S , et al . Synthesis of 5-hydroxymethylfurfural (5-HMF) from fructose over cation exchange resin in a continuous flow reactor [J ] . Chemical Engineering and Processing - Process Intensification , 2019 , 138 : 65 - 72 .

Shen Y , Sun J K , Wang B , et al . Catalytic approaches to the production of furfural and levulinates from lignocelluloses [M ] // Biomass as Renewable Raw Material to Obtain Bioproducts of High-Tech Value . Amsterdam : Elsevier , 2018 : 235 - 269 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

A hybrid modeling and optimization strategy for sequential simulated moving bed

Transition metal-modified MoS2 for catalytic hydrodeoxygenation of lignin-derived phenols to aviation fuel-range arenes

Research progress on the application of low-temperature plasma in biomass gasification to produce hydrogen

Advancements and obstacles in the production of nitrogen-containing bio-based chemicals from chitin biomass

Evolution of surface functional groups in the pyrolysis of lignin with the introduction of polyethylene and transition metals

Related Author

LIU Pengpeng

WANG Zhiguo

LUAN Xiaoli

LIU Fei

WU Zifan

DENG Yacong

CHEN Lu

QIU Yuting

Related Institution

Key Laboratory of Advanced Control for Light Industry Process of the Ministry of Education, Jiangnan University

State Key Laboratory of Heavy Oil Processing, College of Chemistry and Chemical Engineering, China University of Petroleum (East China)

College of Architecture & Environment, Sichuan University

College of Carbon Neutrality Future Technology, Sichuan University

Zaozhuang Shanting District Housing and Urban-Rural Development Bureau

⁰